You are reading a single comment by @Kurai and its replies.
Click here to read the full conversation.
-
There's two distinct options for death certificates: Positive Covid19 and Suspected Covid19 (even if they test negative this can be used). Are they doing postmortem swabs?
Also, if I've read ONS correctly, any pneumonia with a positive Covid19 is being counted in the Covid19 pile. This could include things like aspiration pneumonia and pneumoccal pneumonia.
Yes, they already are adding the deaths outside hospital but they do this separately and the only stat they had for the first few weeks was 'deaths in hospital'.
They're probably relying on the fact that a sudden jump in deaths (by now reporting on all deaths in/out hospital) would be seen as a huge jump and spook many people who just look for a headline number.
At first it was the only number they had, and so they're going to continue using that same number as the primary metric. They should also be publishing the total number of known deaths and reiterating that there are delays in gathering these figures.
The problems with deaths outside hospital are:-
a) Deaths outside hospital take much longer to be reported and filter through to the central stats gathering service (ONS)
b) Deaths outside hospital are less likely to have already been tested, therefore there's also a delay for post-mortem nasal swab tests to check for Covid-19 before this can be formally put on a death certificate
c) There's no guarantee that every death outside hospital will be tested, and there's still a judgement call to be made on whether someone who died who tested positive for Covid-19 actually died from the complications of Covid-19 or whether they were asymptomatic or mildly symptomatic and died from something unrelated.
d) etc...
Even for deaths within hospitals much of that applies, there can be significant delays, here's the data from a week ago showing date of death and date of report of death:-
https://twitter.com/RP131/status/1252586603580731393?s=20
So if a death occurs in hospital the most likely date that it will be properly reported is two days later.
Hospitals lag by two days on average, but then there's a long tail from testing.
Deaths outside hospitals lag by weeks.
Adding the deaths from two weeks ago to the "day's stats" makes for misleading and not very useful stats, especially when vast swathes of the current media just want to headline with a single number.